If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
State insurance mandates for in vitro fertilization are not associated with improving racial and ethnic disparities in utilization and treatment outcomes
Division of Reproductive Endocrinology and Infertility, Department Obstetrics, Gynecology and Reproductive Sciences, Yale School of Medicine, New Haven, CT
Racial and ethnic disparities in utilization and clinical outcomes following fertility care with in vitro fertilization in the United States are well-documented. Given the cost of fertility care, lack of insurance is a barrier to access across all races and ethnicities.
Objective
This study aimed to determine how state insurance mandates are associated with racial and ethnic disparities in in vitro fertilization utilization and clinical outcomes.
Study Design
This was a cohort study using data from the Society for Assisted Reproductive Technology Clinical Outcome Reporting System from 2014 to 2019 for autologous in vitro fertilization cycles. The primary outcomes were utilization—defined as the number of in vitro fertilization cycles per 10,000 reproductive-aged women—and cumulative live birth—defined as the delivery of at least 1 liveborn neonate resulting from a single stimulation cycle and its corresponding fresh or thawed transfers.
Results
Most (72.9%) of the 1,096,539 cycles from 487,191 women occurred in states without an insurance mandate. Although utilization was higher across all racial and ethnic groups in mandated states, the increase in utilization was greatest for non-Hispanic Asian and non-Hispanic White women. For instance, in the most recent study year (2019), the utilization rates for non-Hispanic White women compared with non-Hispanic Black/African American women were 23.5 cycles per 10,000 women higher in nonmandated states and 56.2 cycles per 10,000 women higher in mandated states. There was no significant interaction between race and ethnicity and insurance mandate status on any of the clinical outcomes (all P-values for interaction terms > .05).
Conclusion
Racial and ethnic disparities in utilization of in vitro fertilization and clinical outcomes for autologous cycles persist regardless of state health insurance mandates.
Racial and ethnic disparities in clinical outcomes following fertility care with in vitro fertilization (IVF) have been identified in the United States since the early 2000s and have persisted through the last 2 decades.
Comparison of assisted reproductive technology utilization and outcomes between Caucasian and African American patients in an equal-access-to-care setting.
Multivariate analysis of factors affecting probability of pregnancy and live birth with in vitro fertilization: an analysis of the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System.
Society for Assisted Reproductive Technology Writing Group Report. Trends of racial disparities in assisted reproductive technology outcomes in Black women compared with White women: Society For Assisted Reproductive Technology 1999 and 2000 vs. 2004-2006.
Most recently, a systematic review identified 37 studies through April 2021 that provided consistent evidence of racial and ethnic disparities in access to infertility care and treatment outcomes.
Current estimates from the Centers for Disease Control and Prevention (CDC) indicate that over 19% of nulliparous women aged 15 to 49 years are infertile but only 12% have ever used fertility services.
Recent studies suggest that the prevalence of infertility among Hispanic and Black women is similar to or higher than non-Hispanic White women, but the use of fertility services is highest among non-Hispanic White women and lowest among Black and Hispanic women.
It is expected that IVF utilization will only accelerate in the future, thus highlighting the importance of addressing racial and ethnic disparities in IVF care as a pressing issue.
This study aimed to determine how state insurance mandates are associated with racial and ethnic disparities in in vitro fertilization utilization and clinical outcomes in autologous cycles.
Key findings
The average utilization rates among non-Hispanic Black/African American and Hispanic/Latinx women in mandated states were lower than or similar to the average utilization rate among non-Hispanic White and Asian women in nonmandated states. Non-Hispanic Black/African American, Hispanic/Latinx, and Asian women have lower odds of live birth and clinical pregnancy than non-Hispanic White women regardless of state insurance mandate status.
What does this add to what is known?
Racial and ethnic disparities in in vitro fertilization utilization for autologous cycles have not been mitigated or resolved by state-mandated insurance. Disparities in clinical outcomes persist regardless of state health insurance mandates.
The American Society for Reproductive Medicine reported that the average cost of IVF was $12,400 in 2015, whereas another study estimated the median expense at over $19,000, depending on the region of the United States.
Given the high cost of fertility care, lack of insurance coverage is a barrier to access across all races and ethnicities. In an effort to improve access, 19 states have enacted laws that require private third-party insurers to cover diagnosis or treatment of infertility or fertility care, and of those, 9 require coverage of IVF. At least 6 states considered similar legislation in their 2022 legislative sessions.
Only New York’s mandate requires coverage by its Medicaid program, which went into effect in 2020. Statewide data from Massachusetts found that insurance mandates increased the total utilization of fertility care, but predominantly among those who were already most likely to access it, specifically, White, educated, wealthy women.
This raises important questions about whether state-mandated fertility coverage for IVF exacerbates racial disparities in utilization.
Insurance coverage may also impact cumulative live birth rates. Recent national success rates for a single IVF cycle range between 55% for patients <35 years and 4.3% for patients >42 years.
Hence, many patients undergo multiple treatment cycles before achieving a live birth, and undergoing more treatment cycles incurs additional expenses. A recent retrospective cohort study using 2018 national IVF data concluded that comprehensive state-mandated insurance coverage of IVF services was associated with higher live birth rates per cycle, albeit with a small effect size, among other positive clinical outcomes.
However, they did not have patient-level data and thus could not assess whether the association between comprehensive mandates and clinical outcomes were the same for all races and ethnicities. Another study found in a secondary analysis that states with mandates for IVF had improved cumulative live birth rates among non-Hispanic Black/African American women, whereas no such improvement was observed among non-Hispanic White women.
At a time when more states are considering enacting mandates, our study aimed to provide an up-to-date assessment of the association between state insurance mandates and racial disparities in utilization and treatment outcomes using data from a large validated nationwide IVF registry.
Materials and Methods
This study was approved by the institutional review board of Amherst College and the Society for Assisted Reproductive Technology (SART) Research Committee. The data used for this study were obtained from the SART Clinic Outcome Reporting System (CORS). Data were collected through voluntary submission, verified by SART, and reported to the CDC in compliance with the Fertility Clinic Success Rate and Certification Act of 1992 (Public Law 102-493). In 2018, 86% of all assisted reproductive technology (ART) clinics in the United States were SART members.
The data in SARTCORS are validated annually, with 7% to 10% of clinics receiving on-site visits for chart review based on an algorithm for clinic selection. During each visit, data reported by the clinic were compared with information recorded in patients’ charts. In 2019, 9 out of 11 data fields selected for validation were found to have discrepancy rates of ≤5%.
Utilization was measured as the number of autologous IVF cycles per 10,000 women of reproductive age (18–45 years old). Population estimates by state, age, race, and ethnicity were obtained from the US Census Bureau Population Division.
Utilization rates were computed for each combination of year, state, and race and ethnicity group.
The primary clinical outcome of interest was cumulative live birth, defined as the delivery of at least 1 liveborn neonate resulting from a single stimulation cycle and its corresponding fresh or thawed transfers. The secondary outcomes of interest included live birth per cycle start with intent to transfer, live birth per embryo transfer, clinical pregnancy per cycle start with intent to transfer, clinical pregnancy per embryo transfer, spontaneous abortion per pregnancy, and cycle cancellation. Cycles with a plan to do preimplantation genetic testing or embryo freezing with no intent to transfer were excluded from the “per cycle start with intent to transfer” outcomes given that the lack of transfer in these cycles is owing to prespecified patient choice.
State insurance mandate assessment
Eight states as follows that had insurance mandates in effect for the entirety of time between January 1, 2014 and December 31, 2019 for any coverage of ART services were included in the insurance mandate group: Arkansas, Connecticut, Hawaii, Illinois, Maryland, Massachusetts, New Jersey, and Rhode Island. Delaware was categorized in the insurance mandate group only for 2019, as its mandate went into effect in the middle of 2018. All other states and the District of Columbia (DC) were categorized as having no insurance mandate for coverage of IVF services.
Race and ethnicity assessment
The prompt for reporting patient race and ethnicity data to SART states, “Race/Ethnicity should be ascertained by asking: With which of the following racial/ethnic groups do you, the patient, most closely identify? (Select ALL that apply)” with options for: American Indian or Alaska Native, Asian, Black/African American, Hispanic/Latino, Native Hawaiian/Other Pacific, White, Not Asked, Refused, and Unknown. For this analysis, any cycle with “Hispanic/Latino” selected was categorized as Hispanic/Latinx ethnicity and the remaining groups were for non-Hispanic women: non-Hispanic White, non-Hispanic Black/African American, non-Hispanic Asian, other (which combined American Indian or Alaska Native and Native/Hawaiian/Other Pacific owing to small numbers), or multiple races. Cycles with “Not Asked,” “Refused,” or “Unknown” selected were considered missing patient race and ethnicity information.
Missing data
Multiple imputation by chained equations (MICE) was utilized in the primary analysis. Twenty imputed datasets were created using the MICE package in R.
Classification and regression tree methods were utilized for all imputation models because of their flexibility in fitting interactions and nonlinear relationships and their ability to handle multicollinearity and skewed distributions.
Full R code for the implementation of MICE, including specification of the predictor matrix for each imputation model, can be found at https://github.com/katcorr/SARTCORS_MI. Austin et al
provide a tutorial on multiple imputation in clinical research and can be referenced for further details on how MICE addresses missing data in a principled manner.
Statistical analysis
All statistical analysis was conducted in R version 4.1.0 (R Core Team, 2021).
Utilization rates and their 95% confidence intervals were estimated using generalized estimating equations (GEE) with a Poisson distribution allowing for overdispersion, log link, and an exchangeable correlation structure to account for clustering by clinic state. A separate model was fit for each year, providing the most flexibility in the relationship between mandate, race and ethnicity, and utilization across time. Each model included mandate, race and ethnicity, an interaction term between mandate and race and ethnicity, and an offset term for the population size.
For each clinical outcome, adjusted odds ratios and 95% confidence intervals were generated using GEE with a binary outcome, logit link, and exchangeable correlation structure to account for clustering by clinic state using the 1-step GEE approach for large cluster sizes.
Robust standard errors were used to provide valid standard errors, even in the presence of misspecification of the correlation matrix. Two different adjusted models were fit. The first adjusted model included patient demographics including patient age, patient age squared (to accommodate the quadratic relationship between patient age and the log odds of the outcome), patient body mass index (<18.5 kg/m2, 18.5–24.9 kg/m2, 25.0–9.9 kg/m2, ≥30 kg/m2), maximum follicle stimulating hormone level (≤10 mIU/mL, >10 mIU/mL), anti-Müllerian hormone level (<1 ng/mL, 1–4 ng/mL, >4 ng/mL), previous spontaneous abortion, parous, endometriosis, male factor infertility, tubal factor infertility, ovulatory factor infertility, uterine factor infertility, diminished ovarian reserve, and unexplained infertility. The second adjusted model included all the covariates in the first adjusted model plus the following 3 cycle-level covariates: any use of intracytoplasmic sperm injection (ICSI), number of embryos transferred (0, 1, 2, 3, 4+), and cycle type (fresh or frozen).
Results
Patient and cycle characteristics
A total of 1,096,539 autologous IVF cycles from 487,191 different women were included in the analysis. Most (72.9%) cycles occurred in states that do not have an insurance mandate for coverage of IVF services. The median age at IVF cycle start was 35 years in states without an insurance mandate and 36 years in states with an insurance mandate (Table 1). The distribution of infertility diagnoses is similar between the 2 groups. ICSI was less common in the mandate group (85% vs 77%), and single embryo transfer was slightly more common (58% vs 64% of transfers). A substantial proportion of cycles in the nonmandated and mandated groups were egg retrievals with no intent to transfer owing to planned preimplantation genetic testing (18% and 15%, respectively).
Table 1Patient and cycle characteristics among 1,096,539 in vitro fertilization cycles in the United States, 2014–2019
Characteristic
State insurance mandates coverage of IVF services?
Table A.1 provides a comparison of patient and cycle characteristics by patient race and ethnicity, including cycles missing this information. Approximately one-third (35%) of cycles had unknown patient race and ethnicity. The proportion of cycles occurring in a state with an insurance mandate was slightly higher among cycles from non-Hispanic Black/African American women than non-Hispanic White women (33% vs 27%). The median age at cycle start was slightly higher for non-Hispanic Black/African American women (37 years) and Asian women (36 years) than non-Hispanic White women (35 years). There were notable differences in the distribution of body mass index and number of embryos transferred by patient race and ethnicity (Table A.1). Preimplantation genetic testing was most common among non-Hispanic Asian women (22% of cycles) and least common among non-Hispanic Black/African American women (11% of cycles).
In vitro fertilization utilization analysis
For each race and ethnicity group, the average IVF utilization rate was higher among states with an insurance mandate than states without an insurance mandate (Figure). Among non-Hispanic White women, the utilization rate was between 2.4 and 2.5 times higher in states with insurance mandates than states without insurance mandates, with the lowest increase observed in 2018 (rate ratio [RR], 2.42; 95% confidence interval [CI], 1.65–3.55) and the highest increase observed in 2017 (RR, 2.53; 95% CI, 1.72–3.74). The relative increase in utilization rate between mandated and nonmandated states did not differ significantly for non-Hispanic Black/African American and Hispanic/Latinx women compared with non-Hispanic White women (P values for the interaction terms >.05). However, the absolute differences were larger in mandated states. For instance, in the most recent study year (2019), the utilization rates for non-Hispanic White women compared with non-Hispanic Black/African American women were 23.5 cycles per 10,000 women higher in nonmandated states and 56.2 cycles per 10,000 women higher in mandated states. Similarly, the utilization rates for non-Hispanic White women compared with Hispanic/Latinx women were 27.4 cycles per 10,000 women higher in nonmandated states and 64.2 cycles per 10,000 women higher in mandated states.
FigureNumber of IVF cycles per 10,000 reproductive-aged women by patient race and ethnicity and insurance mandate status, 2014–2019
Similar patterns were observed when only cycles with known patient race and ethnicity were included in computing the utilization rates (Figure A.1).
Clinical in vitro fertilization outcomes analysis
In both the complete-case analysis and the analysis on the multiple-imputed data, there were no statistically significant interactions between state insurance mandate status and patient race and/or ethnicity on any of the clinical outcomes (P-values for interaction terms > .05). This lack of significance means that the associations between patient race and ethnicity and clinical outcomes were the same among states with and without an insurance mandate (Table 2), and likewise, the associations between state insurance mandate and clinical outcomes did not differ across patient race and ethnicity groups (Table A.2).
Table 2Adjusted odds ratios and 95% confidence intervals for the association between patient race and ethnicity and clinical in vitro fertilization outcomes within state insurance mandate status group, 2014–2019
Odds ratios and 95% confidence intervals are from generalized estimating equations with a binary outcome, logit link, and an exchangeable correlation structure to account for clustering by state and are adjusted for the following covariates: patient age, body mass index (<18.5, 18.5–24.9, 25–29.9, 30+, missing), follicle stimulating hormone ≥10, anti-Müllerian hormone (<1, 1–4, >4), any previous spontaneous abortion, parous, and reason for assisted reproductive technology, including any male factor infertility, endometriosis, any tubal factor infertility, uterine factor infertility, ovulatory factor infertility, diminished ovarian reserve, and unexplained infertility.
No
Yes
Cumulative live birth
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
0.84 (0.80–0.88)
0.81 (0.77–0.84)
Non-Hispanic Asian
0.86 (0.82–0.89)
0.86 (0.83–0.90)
Other
0.86 (0.77–0.96)
0.86 (0.74–0.99)
Multiple races
0.88 (0.79–0.97)
0.93 (0.83–1.05)
Hispanic/Latinx
0.93 (0.90–0.96)
0.91 (0.89–0.94)
Live birth among cycle starts
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
0.86 (0.83–0.91)
0.85 (0.81–0.88)
Non-Hispanic Asian
0.91 (0.88–0.94)
0.88 (0.85–0.91)
Other
0.91 (0.82–1.01)
0.96 (0.86–1.07)
Multiple races
0.88 (0.80–0.98)
1.01 (0.91–1.12)
Hispanic/Latinx
0.97 (0.95–1.00)
0.96 (0.92–0.99)
Live birth among embryo transfers
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
0.87 (0.83–0.91)
0.85 (0.82–0.88)
Non-Hispanic Asian
0.92 (0.89–0.95)
0.89 (0.86–0.93)
Other
0.91 (0.82–1.01)
0.95 (0.85–1.07)
Multiple races
0.88 (0.77–1.01)
1.00 (0.90–1.11)
Hispanic/Latinx
0.94 (0.91–0.98)
0.95 (0.91–0.98)
Clinical pregnancy among cycle starts
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
0.90 (0.87–0.94)
0.89 (0.86–0.92)
Non-Hispanic Asian
0.92 (0.90–0.94)
0.89 (0.86–0.92)
Other
0.91 (0.84–0.99)
0.91 (0.79–1.03)
Multiple races
0.91 (0.80–1.02)
1.09 (1.00–1.19)
Hispanic/Latinx
1.00 (0.97–1.04)
0.98 (0.93–1.03)
Clinical pregnancy among embryo transfers
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
0.92 (0.89–0.95)
0.88 (0.86–0.91)
Non-Hispanic Asian
0.93 (0.91–0.96)
0.91 (0.88–0.94)
Other
0.91 (0.84–1.00)
0.89 (0.77–1.02)
Multiple races
0.91 (0.78–1.07)
1.09 (0.99–1.21)
Hispanic/Latinx
0.98 (0.94–1.01)
0.97 (0.92–1.02)
Spontaneous abortion
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
1.19 (1.10–1.28)
1.17 (1.10–1.24)
Non-Hispanic Asian
1.07 (1.01–1.13)
1.11 (1.06–1.18)
Other
1.04 (0.84–1.28)
0.80 (0.60–1.08)
Multiple races
1.13 (0.99–1.30)
1.20 (0.97–1.49)
Hispanic/Latinx
1.09 (1.04–1.15)
1.07 (1.00–1.15)
Cycle cancellation
Non-Hispanic White
1.00 (reference)
1.00 (reference)
Non-Hispanic Black/African American
1.15 (1.07–1.22)
1.25 (1.17–1.33)
Non-Hispanic Asian
0.87 (0.77–0.98)
1.02 (0.94–1.10)
Other
1.18 (0.99–1.41)
1.05 (0.77–1.43)
Multiple races
1.06 (0.94–1.20)
1.04 (0.84–1.28)
Hispanic/Latinx
1.02 (0.98–1.06)
1.09 (1.00–1.19)
IVF, in vitro fertilization.
Correia. State insurance mandates for in vitro fertilization and racial and ethnic disparities. Am J Obstet Gynecol 2023.
a Analyses fit to multiple-imputed data
b Odds ratios and 95% confidence intervals are from generalized estimating equations with a binary outcome, logit link, and an exchangeable correlation structure to account for clustering by state and are adjusted for the following covariates: patient age, body mass index (<18.5, 18.5–24.9, 25–29.9, 30+, missing), follicle stimulating hormone ≥10, anti-Müllerian hormone (<1, 1–4, >4), any previous spontaneous abortion, parous, and reason for assisted reproductive technology, including any male factor infertility, endometriosis, any tubal factor infertility, uterine factor infertility, ovulatory factor infertility, diminished ovarian reserve, and unexplained infertility.
After adjusting for patient demographics, non-Hispanic Black/African American women had significantly lower odds of cumulative live birth per index retrieval than non-Hispanic White women in both nonmandated (adjusted odds ratio [aOR], 0.84; 95% confidence interval [CI], 0.80–0.88) and mandated states (aOR, 0.81; 95% CI, 0.77–0.84). Cycles from non-Hispanic Black/African American women also had lower odds of clinical pregnancy and higher odds of spontaneous abortion and cycle cancellation regardless of state mandate (Table 2).
Cycles from non-Hispanic Asian women also fared worse than cycles from non-Hispanic White women in both nonmandated and mandated states, with 14% lower odds of cumulative live birth (aOR, 0.86; 95% CI, 0.80–0.88 in nonmandated states; aOR, 0.86; 95% CI, 0.83–0.90 in mandated states), lower odds of clinical pregnancy per cycle start, and higher odds of spontaneous abortion (Table 2).
The odds of cumulative live birth were significantly lower among Hispanic/Latinx women than non-Hispanic White women in both nonmandated (aOR, 0.93; 95% CI, 0.90–0.96) and mandated states (aOR, 0.91; 95% CI, 0.89–0.94). Hispanic/Latinx women had similar odds of clinical pregnancy per cycle start and slightly higher odds of spontaneous abortion than non-Hispanic White women (Table 2).
Women of other races and multiracial women had lower odds of live birth and clinical pregnancy. However, these estimates were less stable, with more variability and wider confidence intervals because of the smaller group sizes.
ORs and 95% CIs from main-effects only models (ie, models without interaction terms between mandate and race and ethnicity) are summarized in Table A.3. Additional adjustment for ICSI, number of embryos transferred, and cycle type (fresh or frozen) yielded similar results (Table A.3). The complete-case analysis yielded stronger associations in each case (Table A.3).
Comment
Principal findings
Utilization was higher across all racial and ethnic groups in mandated states, but the increase in utilization was greatest for non-Hispanic Asian and non-Hispanic White women. Racial disparities in utilization not only persisted regardless of mandate but were greater in mandated states. The current study also demonstrated decreased prognosis for successful reproductive outcomes from IVF among non-Hispanic Black, non-Hispanic Asian, and Hispanic/Latinx women than non-Hispanic White women, regardless of state mandate status, as measured by cumulative live birth per index retrieval, live birth per cycle start and per embryo transfer, and spontaneous abortion per pregnancy.
Results in the context of what is known
Previous studies that have assessed IVF utilization by race and ethnicity were limited either to analyzing survey data from a single clinic in 2002,
Our updated and more comprehensive assessment of over a million IVF cycles from almost half a million women during the most recent available 6 years demonstrate that IVF utilization is consistent with these older studies that found no evidence that mandates mitigate racial and ethnic disparities.
Previous research on the association between state mandates and clinical outcomes by race and ethnicity is even more limited. A secondary analysis in one study suggested that state mandates may be associated with increased cumulative live birth rates for Black/African American women but not for White women.
Our study found no association between state mandate and cumulative live birth for any race and ethnicity group. The apparent inconsistency in results could be explained by the fact that the previous analysis was unadjusted and found a relatively small difference, whereas our analysis has adjusted for relevant patient and cycle-level confounders.
Clinical implications
There are several explanations for why state insurance mandates could improve utilization across racial and ethnic groups while still perpetuating or exacerbating disparities in utilization and not addressing disparities in clinical outcomes. First, no state’s mandate required its Medicaid program to cover infertility care during the study period. Owing to historic precedents, Black and Hispanic/Latinx women are more likely to participate in Medicaid for their health insurance. Among adults 18 to 64 years old in 2019, 25% of White adults and 22% of Asian adults were uninsured or had health coverage through Medicaid vs 42% of Black adults and 46% of Hispanic adults.
Second, state insurance mandates only govern particular types of insurance plans. Under federal law, states cannot regulate the largest, self-insured employers. In addition, systemic racism likely plays a role in deterring some patients from seeking care. Systemic racism manifests in a lack of providers of color, unconscious provider bias, and models of care that do not sufficiently take social determinants of health into account.
Fertility and Sterility Systemic racism exists in reproductive endocrinology and infertility: we are part of the problem. Fertility and sterility dialog.
The main impediment to research on racial and ethnic disparities in IVF right now is missing data. Improving data collection practices around race and ethnicity
is critical so that future studies that utilize national IVF databases have complete and accurate data to continue tracking racial and ethnic disparities and how disparities may or may not be associated with state mandates in the future. At the same time, race and ethnicity are social constructs, without biological meaning, and serve as proxies for other factors such as racism and cultural and socioeconomic factors.
Future multidisciplinary research efforts will be necessary to develop a deeper understanding of the root causes of these disparities to inform policy and public health interventions. Ongoing research efforts in other obstetrics and gynecology subfields can serve as examples for the IVF field.
This is the most comprehensive study of over 1 million autologous IVF cycles using a large, validated US national IVF database that has examined racial and ethnic disparities in utilization and treatment outcomes as a function of the presence or absence of state insurance mandates. There are many strengths to our study, including the use of an up-to-date, large, comprehensive, and validated nationwide database. The linkage of embryo transfers from a specific index retrieval allowed for the calculation of cumulative live birth rates—the most contemporary and meaningful metric of IVF treatment outcome. Furthermore, owing to the large sample size, we were able to compare 6 different race and ethnicity groups, which is more informative and inclusive than the White/non-White dichotomy that smaller studies are often limited to investigating.
Our study also has some acknowledged limitations, including missing data, with one-third of cycles having unknown patient race and ethnicity. To minimize bias and increase precision, we used a principled approach to handle the missing information by implementing MICE.
Although the assumptions for MICE are likely correct in this context, they cannot be definitively confirmed. We also performed sensitivity analyses, which consistently provided the same general findings and conclusions.
Conclusions
Our study shows that though existing state insurance mandates laudably improve access for all groups, other or additional mechanisms are critical to address racial and ethnic disparities. Although state insurance mandates may be necessary to allow affordability and general access, they do not seem to be sufficient in their present form to result in narrowing or creating equal access to or outcomes from IVF.
Acknowledgments
The authors thank SART for the dataset, and all SART members for providing clinical information to the SART CORS database for use by patients and researchers. Without the efforts of SART members, this research would not have been possible.
Comparison of assisted reproductive technology utilization and outcomes between Caucasian and African American patients in an equal-access-to-care setting.
Multivariate analysis of factors affecting probability of pregnancy and live birth with in vitro fertilization: an analysis of the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System.
Society for Assisted Reproductive Technology Writing Group Report. Trends of racial disparities in assisted reproductive technology outcomes in Black women compared with White women: Society For Assisted Reproductive Technology 1999 and 2000 vs. 2004-2006.
This study utilized funding from the Ferring Innovation Grant for Racial Equality in Reproductive Medicine and Maternal Health (DBS). The funder was not involved in the collection, analysis, or interpretation of data; writing the report; or deciding to submit the article for publication.
Cite this article as: Correia K, Kraschel K, Seifer D. State insurance mandates for in vitro fertilization are not associated with improving racial and ethnic disparities in utilization and treatment outcomes. Am J Obstet Gynecol 2023;228:313.e1-8.
02158-5&id=gr1.jpg){kind=link}